Slow TF down

Lighten up on speed of making the product.

We worked with an organisation who spent millions with one of the BigC consultancies  – I won’t name names but it started with a D – in order to try to improve their time to deliver product to clients. The consultancy focused on their factory and optimised it so that they could make a product in hours instead of days. When we looked at the value stream, it still took months to fulfil a customer order. When we showed the client how they had spent all that money to optimise what wasn’t the problem, they weren’t very happy.

The same is true of IT. I did some work with the TaskTop product, and the biggest thing I learned was that single-digit flow efficiencies are pretty typical of software development. “From a-ha to ka-ching”, from request to fulfil, we spend maybe only 5% of the time actually working on the feature.

One way to see this is to imagine that a manager somewhere in your organisation needed a “hello world” program. If the actual work itself was only a single line of code, how long would it take to propose the app, have it approved, funded, prioritized, assigned for work, developed, tested, deployed to production, supported, trained, documented, then released to the users?  Never mind if it also needs to be priced, packaged, marketed and sold.

Even if it’s ten thousand lines, whether that code is developed by 10x ninjas or me (who hasn’t coded in three decades) won’t make a discernable difference to how soon it sees the light of day in most companies, with their 5% value flow efficiency.

So why do we make such a big deal about fast work?

Firstly because we are all overexcited about competitiveness.

And secondly because we are all overexcited about efficiency.


Competitiveness first.  The percentage of organisations for whom speed to market of new products is a survival imperative is in fact small. Even where it is imperative, it only affects those involved in the new. For most people in most organisations, at least equal priorities are sustainability of work, values and ethics, adaptability and resilience, quality and improvement, and morale.

Speed is interesting, as one amongst several factors that give us agility. Agility is not about speed to deliver. It is about speed to change what we deliver and how we deliver it, in response to changing circumstances. Speed contributes to agility only in so much as it makes each iteration quicker so we have more opportunities – and sooner – to reflect and adjust.

So speed of iteration is useful, but not because we are racing to market. Only those who sell things to the general public live or die on a few months. Most of us aren’t in consumer retail, and most of those in consumer retail aren’t making new products. Even in retail, it depends on the product. Mostly, it’s only digital consumer product. Even then it depends on the industry e.g. banks don’t care (they say they do but their systems say otherwise). That’s also related to size. If the product is your only product, that matters more than to a global corporation. It also depends on the country. The USA is lost in its hyper-competitive dog-eating mindset and has suckered the online world into it, but not all the planet is so aggressive. 

That all adds up to an entrepreneurial startup culture which pervades IT social media and writing, but mostly it’s unnecessarily urgent.  It’s nice to be quick but it’s not gonna wipe you out if it takes a year.


Once we get over the fear and loathing of competitive thinking, we must also see that productivity/efficiency is only one goal, often the wrong one.

Many readers will be familiar with the Cynefin model. When we look at Cynefin, it’s tempting to suggest that in Clear or Complicated work, we can optimise for efficiency, with a simple linear model. We must remember that Cynefin is situational, and the situation can change at any moment. The collapse of Just In Time supply chains during the pandemic exposed the danger.

Optimising for efficiency not agility is a mistake in a VUCA world. My favourite quote on this is from General Stanley McChrystal:

“The 21st Century is a different game with different rules… The pursuit of efficiency was once a laudable goal, but being effective in today’s world is less a question of optimizing for a known (and relatively stable) set of variables than responsiveness to a constantly shifting environment. Adaptability, not efficiency, must become our central competence.”

We should optimise for adaptability everywhere. Adaptability comes from a combination of agility and resilience. All our teams, our systems, need to be able to work in an agile way at the drop of a hat, to respond to changing circumstance. Within that, there is always room to optimise flow, in the knowledge that the flow could change to something different at any time, so don’t over-invest.  And with the constraint that the improvement of flow must never reduce our capacity for agility.

My favorite example is auto manufacturers who once dreamed of humanless factories until they realised it’s not flexible enough. They’re not Henry Ford cranking out identical black cars.

As we make work Clear and defined, our system must always be ready to drop into interactive exploratory mode at any time.

(I accidentally created the mechanism for this with my Standard+Case model .)


There is something more profound at stake here than even competitiveness or adaptability: humanity.

Humans and human systems are not machines and cannot be managed in a mechanical way. Attempting to reduce such things to numbers is actively harmful by oversimplifying that which cannot be simplified without losing much of its value. This is the problem with Taylorism and Scientific Management.

Donella Meadows speaks of “dancing with the system”. Many – perhaps most – sense-making tools in a complex system don’t give us numbers, they give us narrative and consensus. There are no dials on a human.

You don’t NEED numeric measurement to see where your star is. It merely makes you more efficient and effective at tracking it.

Measurement actively reduces the value of a human system. Maybe it is “better” to be intentional/directed/controlled in the pursuit of profit or dogma, but we lose the diversity and humanity of organic systems. Reality is organic, we only impose simple linear models on it, to our ultimate cost. Adaptability is more important to survival than performance.